MULTIPARAMETRIC MRI FOR PROSTATE CANCER

Multiparametric MRI (MP-MRI) administered as a triage test before a man's first transrectal ultrasound (TRUS) biopsy could identify at least one quarter of all patients who could safety avoid having a biopsy altogether, and it might also improve the detection of clinically significant prostate cancer, British researchers conclude from the PROMIS study.

"Our results show that MP-MRI should be used before [TRUS] biopsy," lead author, Hashim Ahmed, MD, University College London, United Kingdom, said in a statement.

"Biopsies will still be needed if an MP-MRI scan shows suspected cancer, but the scan could help to guide the biopsy so that fewer and better biopsies are taken," Dr Ahmed added.

The study was published online January 19 in The Lancet.

PROMIS was a prospective, paired-cohort study involving men with a clinical suspicion of prostate cancer but no history of prostate biopsy.

The 576 men in the current analysis included those with an elevated serum prostate-specific antigen level of up to 15 ng/mL within the last 3 months or who had a suspicious digital rectal exam. Men suspected of having confined stage T2 or lower disease on rectal exam or with a positive family history were also included.

"Our primary objectives were to establish the proportion of men who could safely avoid biopsy and the proportion of men correctly identified by MP-MRI to have clinically significant prostate cancer," Dr Ahmed and colleagues observe.

They also compared the accuracy of each of the two approaches in terms of sensitivity, specificity, and positive and negative predictive value, as well as the presence of clinically significant prostate cancer using template prostate mapping (TPM) biopsy as a reference. As Dr Ahmed explained, TPM biopsy samples the entire prostate and has an estimated 95% sensitivity for the detection of clinically significant prostate cancer.

Of the 576 men who underwent all 3 tests, TMP biopsy detected cancer in 71% of the cohort, 40% of whom were felt to have clinically significant cancer according to study definitions of significant disease.

"For clinically significant cancer, MR-MRI was more sensitive [at] 93%...than TRUST biopsy [at] 48%," investigators report. As the investigators observed, MR-MRI provides information about the size of a cancer, how densely packed its cells are, and how well connected to the bloodstream it is, which helps the test differentiate between aggressive and benign tumors.

However, sensitivity and specificity values were significantly different between the two approaches and were not always in favor of MP-MRI.

Table. Prevalence of Clinically Significant Cancer

Test Characteristic	MP-MRI (%)	TRUS Biopsy (%)	P Value
Sensitivity	93	48	<.0001
Specificity	41	96	<.0001
Positive predictive value	69	98	<.0001
Negative predictive value	89	74	<.0001

As Dr Ahmed and colleagues point out, the 89% negative predictive value seen with the MP-MRI triage approach is "reassuring" in that it implies that there is, indeed, no clinically significant cancer in the prostate. It has been previously reported that MP-MRI often "misses" low-risk prostate cancer and hones in on the more clinically significant lesions reference  — a clear advantage in a diagnostic test for a cancer such as prostate cancer.

Worst- vs Best-Case Scenario

The investigators went on to model what would happen if clinicians used MP-MRI as a triage test instead of standard TRUS biopsy, and then sent only men with a suspicious MP-MRI score on to biopsy. They assumed a "worst-case scenario" where TRUS biopsies would automatically be done and a "best-case scenario" where biopsies would be guided by MP-MRI findings.

"For both these scenarios...27% of 576 men would avoid a primary biopsy," Dr Ahmed and colleagues reports. Investigators also predicted that 5% fewer absolute cases of clinically insignificant cancers for the overall cohort would be detected in the worst-case scenario, whereas with the best-case scenario, there would be a 21% absolute reduction in clinically insignificant cancers for the overall cohort.

A total of 44 serious adverse events were documented during the larger PROMIS trial, with 1% of the 576 men involved in the current analysis developing sepsis secondary to urinary tract infection. There was also a 10% urinary retention rate. All of the adverse events were due to the biopsy itself and not to MP-MRI, according to a statement.

"Our current biopsy test can be inaccurate because tissue samples are taken at random [and] this means [TRUS biopsy] cannot confirm whether a cancer is aggressive or not and it can miss aggressive cancers that are actually there," Dr Ahmed noted in the statement.

"While combining the two tests gives better results than biopsy alone, this is still not 100% accurate so it would be important that men would still be monitored after their MP-MRI scan," he added.

"Amazingly Complicated" Study 

Asked by Medscape Medical News to comment on the study, Marc Garnick, MD, professor of medicine, Harvard Medical School and the Beth Israel Deaconess Medical Center, Boston, Massachusetts, felt that the study was actually an "amazingly complicated" effort in terms of what investigators actually did to confirm their findings.

"They ended up doing what we call mapping biopsies, which is our gold standard, against which they could compare the other two diagnostic modalities," Dr Garnick said. "That in itself is a huge effort," he added.

As Dr Garnick noted, centers in the United States will sometimes do a mapping biopsy if they are trying to identify and create a three-dimensional volumetric map of where a cancer in the prostate gland might be.

"The fact that these investigators were able to accomplish this procedure [in so many men] is really amazing," Dr Garnick emphasized.

In his own center, Dr Garnick and colleagues are keenly interested in prostate MRI, where they specifically look at the prostate with the help of an endorectal coil, initially developed by the head of radiology in their own department. As Dr Garnick explained, the endorectal coil provides a clearer signal of the MRI because it's coming from the coil itself, which is tucked up close to the prostate.

"It's not clear whether this group of investigators used an endorectal coil or not, but if they didn't, they probably could have improved upon their results had a coil been used," Dr Garnick said.

"This is a really important piece of information and the authors need to be congratulated for doing such a study," he said, adding that over time, "I think [the MP-MRI] is going to be an important component in the decision making about identifying cancers that hopefully need to be treated, as opposed to cancers that only need to be diagnosed and then you can start asking the question, does treatment really make a difference in any of these cancers?"

Dr Ahmed receives funding from Sonacare Medical, Sophiris, and Trod Medical for clinical trials. Dr Garnick is editor-in-chief of the Harvard Medical School annual report on prostate disease and website. 

Lancet. Published online January 19, 2017. Full text